linux-kernel - Re: [PATCHv3 2/2] serial: Add driver for the Altera UART

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date:	Tue, 30 Mar 2010 08:18:47 +0800
From:	Thomas Chou <thomas@...ron.com.tw>
To:	Tobias Klauser <tklauser@...tanz.ch>
CC:	Andrew Morton <akpm@...ux-foundation.org>,
	linux-serial@...r.kernel.org, gregkh@...e.de,
	nios2-dev@...c.et.ntust.edu.tw, linux-kernel@...r.kernel.org,
	Alan Cox <alan@...rguk.ukuu.org.uk>
Subject: Re: [PATCHv3 2/2] serial: Add driver for the Altera UART

On 03/29/2010 09:29 PM, Tobias Klauser wrote:
> On 2010-03-25 at 01:54:09 +0100, Thomas Chou<thomas@...ron.com.tw>  wrote:
>    
>> On 03/25/2010 12:24 AM, Tobias Klauser wrote:
>>      
>>> On 2010-03-24 at 12:05:27 +0100, Andrew Morton<akpm@...ux-foundation.org>   wrote:
>>>
>>>        
>>>> On Wed, 24 Mar 2010 07:47:47 +0100 Tobias Klauser<tklauser@...tanz.ch>   wrote:
>>>>
>>>>          
>>>>> On 2010-03-23 at 22:54:59 +0100, Andrew Morton<akpm@...ux-foundation.org>   wrote:
>>>>>
>>>>>            
>>>>>> On Fri,  5 Mar 2010 17:52:23 +0100
>>>>>> Tobias Klauser<tklauser@...tanz.ch>   wrote:
>>>>>>
>>>>>>
>>>>>>              
>>>>>>> +	sigs |= (altera_uart_getppdcd(port->line) ? TIOCM_CD : 0);
>>>>>>> +	sigs |= (altera_uart_getppdtr(port->line) ? TIOCM_DTR : 0);
>>>>>>>
>>>>>>>                
>>>>>> We seem to be missing a few things here.
>>>>>>
>>>>>> drivers/serial/altera_uart.c: In function 'altera_uart_get_mctrl':
>>>>>> drivers/serial/altera_uart.c:100: error: implicit declaration of function 'altera_uart_getppdcd'
>>>>>> drivers/serial/altera_uart.c:101: error: implicit declaration of function 'altera_uart_getppdtr'
>>>>>> drivers/serial/altera_uart.c: In function 'altera_uart_set_mctrl':
>>>>>> drivers/serial/altera_uart.c:114: error: implicit declaration of function 'altera_uart_setppdtr'
>>>>>>
>>>>>>              
>>>>> These should usually be declared in a board specific header. There were
>>>>> compatibility macros in altera_uart.c which defined them to NOPs in case
>>>>> the board header did not properly define them. But I remove them as per
>>>>> request by Alan Cox (Message-ID: 20100301181920.3952c3e7@...rguk.ukuu.org.uk).
>>>>>
>>>>> Should we add them again (maybe to altera_uart.h)? Or would it be better
>>>>> to define a config symbol which is set in the board specific Kconfig and
>>>>> altera_uart depends on it?
>>>>>
>>>>>            
>>>> I guess the latter.
>>>>
>>>> There should have been a real implementation of these in the patchset -
>>>> otherwise the code can't be used or tested.  Confused.
>>>>
>>>>          
>>> Sorry for the confusion.
>>>
>>> The last patchset I submitted (with the functions removed from
>>> altera_uart.c) was tested on a local branch, where I added the macros to
>>> a global board specific header. I didn't include that one in the patch.
>>>
>>> After looking at the code and it's history a bit closer (and also on the
>>> nios2 specific part) I realised that this macro was probably added
>>> because the driver was originally based on drivers/serial/mcf.c (the
>>> macros are present there too).
>>>
>>> Also there are currently no board configurations known to me that define
>>> these macros. So I'd suggest to remove the usage of these macros
>>> alltogether. We could still add them again (to the board specific part
>>> and with the config option then) in case there will be a board
>>> configuration implementing DTR/DCD lines on GPIOs.
>>>
>>> Could anyone on nios2-dev verify that there are currently no such board
>>> configurations?
>>>
>>> I'd remove the usage of the macros then and post an updated patch.
>>>
>>>        
>> Maybe we can add pointers to functions for the DCD/DTR in the struct
>> altera_uart_platform_uart. Then board config file can define them if
>> they implement these pins, NULL/0 otherwise.
>>      
> Unfortunately we then need to define them in struct altera_uart too and
> set them in the probe function using the platform data.
>
> For me that seems like a bit too much overhead just to support the
> possibility of someone defining these functions in the future. I'd
> prefer to get rid of the macros now for the mainline submission and then
> add it later if someone would actually use these DCD/DTR get/set
> functions. Hope that's OK for you too, Thomas.
>    
Hi Tobias,

OK. Please drop them. That would be easier.

Best regards,
Thomas
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/